基于形态学和Otsu的固体火箭发动机CT缺陷三维分割  

3D CT Defects Segmentation of Solid Rocket Motor Based on the Morphology and Otsu

作　　者：陈培兴[1,2] 王明泉[1,2] 李世虎[1,2] 侯慧玲[1,2] 王玉[1,2] 

机构地区：[1]中北大学仪器科学与动态测试教育部重点实验室,山西太原030051 [2]中北大学信息与通信工程学院,山西太原030051

出　　处：《图学学报》2015年第4期581-586,共6页Journal of Graphics

基　　金：国家自然科学基金资助项目(61171177);山西省青年科技研究基金资助项目(2012021011-1)

摘　　要：为了能够准确判别固体火箭发动机内部缺陷的性质和可能对发动机造成的危害,需要从三维空间的角度来观察分析。在传统的缺陷分析中,主要是通过观察CT的二维切片序列图像以及对二维图像的主观分析去发现缺陷体。为了对缺陷体进行更为准确、立体地分析,提出对固体火箭发动机工业CT三维体数据进行处理。首先,通过结合形态学和Otsu阈值分割方法对缺陷进行分割和提取;然后,重构出缺陷体三维体数据;最后,对体数据进行三维可视化显示。实验结果表明,该方法能有效、准确地分割和提取固体火箭发动机三维CT图像缺陷,具有较强的鲁棒性。In order to accurately judge the nature of the defect on solid rocket motor and the possibledamage to the engine, it is needed to observe and analyze problems in the three dimensional space.However, the defects is usually determined by observing the two-dimensional slice CT imagesequences and judging by subjective analysis of two-dimensional images in traditional defect analysis,and it requires abundant experience of the user. In order to analyze the defect structure moreaccurately and stereoscopically, the solid rocket motor industrial CT 3D volume data is used to extractdefects by using of the combination of morphological and Otsu threshold segmentation method tosegment. The 3D volume data of the defect structure is reconstructed herein. Finally, 3D visualizationtools are used to display the volume data. The experiment results indicate that this method is effectiveand accurate to segment and extract the solid rocket motor 3D CT image defects, and it has prettystrong robustness.

关 键 词：固体火箭发动机 三维图像分割 缺陷提取 形态学滤波 三维重建 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]